Understanding Urban Inequalities in Bangladesh: A prerequisite for achieving Vision 2021
1
2
Understanding Urban Inequalities in Bangladesh: A prerequisite for achieving Vision 2021
A study based on the results of the 2009 Multiple Indicator Cluster Survey
UNICEF Bangladesh November 2010
3
Acknowledgements This paper was prepared by Dirk Westhof, demographer, with contributions from Carel de Rooy, Siping Wang and Deqa Ibrahim Musa of UNICEF Bangladesh.
© United Nations Children’s Fund (UNICEF), Bangladesh All UNICEF materials are protected by copyright, including text, photographs, images and videotapes. Permission to reproduce any part of this publication must be requested from: Planning, Monitoring and Evaluation Section United Nations Children’s Fund 1, Minto Road, Dhaka-1000 Bangladesh Telephone: 88 02 8852266 Email:
[email protected] Website: www.unicef.org.bd The views expressed in this report are those of the consultant and do not necessarily represent those of UNICEF. ISBN: 984-70292-0022-3
4
Cover Photo: UNICEF/2010/Ahsan Khan Design & layout: Syed Makhlesur Rahman/UNICEF Printing: Evergreen Printing & Packaging
Executive Summary
© UNICEF/2010/Ahsan Khan
Conventional wisdom in Bangladesh says that investing in urban slums will attract more rural migrants. Bangladesh is not alone. Brazil in the past adopted policies that discriminated against urban settlements by the poor, but eventually changed course. Ample evidence shows that people in Bangladesh migrate from rural to urban areas in search of economic opportunities, not in search of basic social services. Indeed, such services are mostly nonexistent in slums, and when they are, they are provided mostly through unscrupulous middlemen using exploitative means – at an extremely high cost to slum dwellers. Data from the 2009 Multiple Indicator Cluster Survey (MICS) in Bangladesh make it clear that conditions in slum areas are much worse than those in most rural areas, even in regard to service delivery-type indicators such as secondary education attendance rates and skilled attendance at birth. The natural trend towards urbanization cannot simply be halted or reversed. In fact, it is an important aspect of the country’s development process and inherent to its evolution from a low income country to a middle-income country. It is also fundamental to the realization of “Vision 2021”, which foresees economic growth that provides an average per capita income of USD2,000 a year. Taking its cue from other countries, Bangladesh should consider this as a positive trend and act proactively by developing a socially inclusive urban development strategy in its sixth Five-Year National Development Plan. This would enable a rapidly growing slum population to more effectively contribute to the acceleration of economic growth and poverty reduction, instead of becoming an ever-increasing burden and obstacle to the development. Evidence from India, China and Brazil indicates very clearly that efforts to ease inequalities generate larger dividends for poverty reduction than a more conventional focus on economic growth. It is time to accept this new paradigm and start to invest in human capital development in the slums of Bangladesh.
5
Introduction This report challenges the widely held belief that the situation of urban dwellers in Bangladesh is generally better than those living in rural areas. In fact, data drawn from a 2009 Multiple Indicator Cluster Study (MICS) clearly show that living conditions in urban slums are often appalling and, in fact, much worse than those in most rural areas.
© UNICEF/2009/Shehzad Noorani
development strategy, since, historically, most development programmes in Bangladesh have focused on rural areas. Evidence from countries including Brazil, China and India show that reducing inequalities yields larger dividends for poverty reduction than a more conventional focus on economic growth, and is a ! " ! development. Dhaka is one of the fastest growing mega-cities in the world, with slum populations seemingly outpacing the growth of other urban areas. Still, the country has no comprehensive policy on urbanization or urban poverty reduction. In the past it was widely believed that people migrated to the cities in search of better services. In fact, when asked about their reasons for migrating, most Bangladeshis cite economic factors.1 And the areas in which migrants initially choose to settle – slums – are those with the lowest levels of services. Denying basic services to slum dwellers, # ! $ % ! "2 1
World Bank, ‘Improving Living Conditions for the Urban Poor’, Dhaka, June 2007. See: Martine, George and Gordon McGranahan, ‘Brazil’s Early Urban Transition: What can it teach other urbanizing countries?’, International Institute for Environment and Development and United Nations Population Fund, August 2010.
2
6
The Current Study
© UNICEF/2009/Kiron
Several studies have been undertaken on the situation of the urban poor in Bangladesh. These studies generally focused on the poorest 20 per cent of the urban population, rather than the situation of " #! '* # reveals large inequalities between slum and other populations – both urban and rural. The picture that results is sobering. The data in this report are drawn from slums areas in four large cities (Dhaka, Chittagong, Khulna and Rajshahi), involving 134 clusters of 20 households each. The report looks at several social indicators and compares them with those of people living in nonslum urban and rural areas.
Background Bangladesh is the most densely populated country in the world (excluding some city-states), with an average of 1,050 people per square kilometre in 2005 (in contrast, the Netherlands has almost 400 people per square kilometre, and Brazil, 20). The 2009 MICS data showed that, in contrast to most countries, differences in coverage in social services are unusually small between urban and rural areas in Bangladesh, and for many indicators they are non-existent. However, when statistics for slum dwellers and other populations are disaggregated out, the situation changes dramatically. For example, while the national average for net attendance in secondary education is 49 per cent (48 per cent in rural areas and 53 per cent in urban areas), it is only 18 per cent in urban slums. The situation is similar for most other social indicators, including skilled !# + ! # better-off urban areas outperforming them both. Net attendance in secondary education is 64 per cent in Khulna City Corporation, for example, and just 14 per cent in Dhaka’s slums. Despite their proximity to metropolitan areas where such services are available, access for slum dwellers is denied. Some improvements in the social sector have been made since the 2006 MICS. However, when it comes to meeting the Millennium Development Goal targets, slums are ;! !% # or upazila, levels). The fact that parts of cities (in this case, city corporations) are in a privileged position is ! "< # a new one for Bangladesh. Normally, based on physical proximity alone, slum dwellers should have greater access to basic social services than isolated rural areas. But this is not the case in Bangladesh.
7
Key concepts Rapid urbanization = ! > ! # is readily apparent that the country is experiencing one of the most rapid urbanization O# Q + "' agree that urban populations in Bangladesh have grown from 5 per cent in 1971 to a 27 per cent in 2008,3 suggesting that approximately 41 million people are currently living in urban areas. Nevertheless, Bangladesh remains a mostly agrarian country. The consensus estimate is that the urban population will reach 50 million in 2015, representing just less than a third (30 per cent) of the total population. The contribution of the urban sector to the national gross domestic product (GDP) has also increased rapidly, from 26 per cent of GDP in 1972 to more than 50 per cent by 2005.4 In 2000, industry represented 20 per cent of all employment in Dhaka; half of that was in the garment industry. Twenty eight per cent of female employment is in the garment industry;5 this in itself is an important factor in female empowerment.
Urban populations > ! + " Z Participation Research Centre (PPRC)6#$ involves a particular population density per square kilometre, a set minimum population size, and over 50 per cent of the labour force in non-agricultural occupations. > ! ! size: Mega-city (Dhaka): A statistical metropolitan area (SMA) with more than 5 million
inhabitants. The population of Dhaka was estimated at 10.7 million in 2001 A statistical metropolitan area: A city corporation and adjacent areas having urban
characteristics (Chittagong, Khulna and Rajshahi) Municipality (pourashava): Area declared by the Local Government Division of
Ministry of Local Government, Rural Development and Co-operatives. Other urban areas: Subdistrict (upazila) headquarters with urban characteristics.
An increase in the number of urban centres has occurred mainly as a result of the " During the current decade, the number of pourashavas jumped from 213 in 2001 to 309 \]]^" ! pourashava; this is up to local governments that might seek (or not seek) an enhanced status. \]]^'* !
_`{ # 2) city corporations (non-slum areas) and 3) slums. The non-city corporation areas were excluded from the urban category and, for this reason, the proportion of the ‘urban’ 3 Tavares-Goodman, Nadia, ‘Review of UNICEF Interventions in the Urban Sector and Potential Future Interventions’, UNICEF Bangladesh, February 2010. 4 Ibid. 5 Islam and Mustafa, ‘Poverty Impact Assessment’, January 2006, UNDP. 6 Power and Participation Research Centre, June 2010.
8
population as a share of the total population in the sampling framework is 20 per cent and not 27 per cent. The ‘slums’ sample frame was the same as for the 2006 MICS and was based on the 2006 World Food Programme report, ‘Study of Household Food Security in Urban Slum Areas of Bangladesh’. For the 2009 MICS, 134 clusters (probability proportional to size) in the slums were visited. In each cluster, at least 20 households were interviewed. Slums7 According to the 2005 Census of Urban Slums, the total slum population in the six largest cities of Bangladesh was 5.4 million compared to a total urban population of roughly 15.6 million. However, there is no typical slum; the differences among such settlements !
"* are poor live in slums.8 Some estimates of the total slum population in 2010 are as high as 10 million. On the other hand, a World Bank study counted 1,925 slums in Dhaka comprising 275,000 households (1.5 million people). Because of several uncertainties, it is preferable to deal in ranges. The population of slums in 2008 was estimated at between six and seven million people, 30 per cent of the metropolitan population and about 15 per cent of the overall urban population. In other words, the population of slums is about 5 per cent of the total population or about 7 million people in 2010. Characteristics of slums (from the 2005 Census of Urban Slums): Predominantly poor housing Very high population density and room crowding (more than 1,000 persons per acre) Very poor environmental services Very low socio-economic status for the majority of residents Lack of security of tenure (permanent threat of eviction)
Urban slums in Bangladesh grew much faster than the overall rate of urbanization. In most cases, slums lack even the most basic amenities associated with urban life, such as running water, sewage systems, latrines, waste disposal services and electricity. Slum dwellers also have restricted access to basic social services, such as health clinics and schools.9
© UNICEF/2010/Ahsan Khan
Poor governance.9
7 This report uses the term ‘slum’ for what can also be termed as ‘informal settlement’, ‘squatter settlement’, ‘pavement dwellings’ and ‘shanties’. 8 Power and Participation Research Centre, June 2010. 9 Tavares-Goodman, February 2010.
9
Migration Migration is the main driver of urban growth. The main reasons for migration are: Economic; ‘push and pull factors’ (55 per cent) Environmental; mostly river erosion (27 per cent) | ; { "10 (10 per cent)
Migration is fuelled by extreme rural poverty, landlessness and large urban-rural wage "'Q + _ search of jobs (employment in services and industry) and better lives.11 Migrants normally end up in illegal settlements on precarious land with major environmental problems. These areas have few – usually low-quality – services, and rents offered are at high prices through middlemen (mastaans) effectively and conveniently positioned between huge demand for and limited offer of basic services. Slum evictions are a constant threat and occur without a clear resettlement plan.
© UNICEF/2009/Shehzad Noorani
The level of services provided in slums is much lower than in most rural areas. Access to better services is almost never given as a reason for migration.
A vicious circle Lack of tenure security is a key characteristic of informal urban settlements. Informal settlements are thus deprived of support and services from the formal public and nonpublic sectors. Some support comes from non-governmental organizations (NGOs) and the United Nations, but these agencies allocate the majority of funding for poverty alleviation to rural areas. The insecurity of tenure and its implicit potential for politicization serve as deterrents when it comes to allocating funds for the urban poor (even for the Grameen Bank). The government perceives that granting tenure security would encourage more settlers and induce migration from rural areas. Thus, the lack of secure tenure perpetuates the downward spiral of poverty in slum areas.12 10
Power and Participation Research Centre, June 2010. World Bank, June 2007. 12 Ahmed, K. Iftekhar: ‘Urban Poor Housing in Bangladesh and Potential Role of Asian Coalition for Housing Rights’ (ACHR), May 2007, ACHR. 11
10
Analysis of Data Urban-rural differentials
countries ! % (U5MR) among selected Asian countries, as shown in Table 1. Table 1: }!% % in selected Asian countries Per 1,000 live births
Ranked by differences Total U5MR
Urban U5MR
Rural U5MR
Rural % higher
Bangladesh (2007)
74
63
77
22%
Bangladesh (2009)
64
53
66
25%
Pakistan (2007
93
78
100
28%
Cambodia (2005)
106
76
111
46%
India (2007)
74
52
82
58%
Indonesia (2007)
51
38
60
58%
Philippines (2008)
37
28
46
64%
Nepal (2006)
79
47
84
79%
Based upon DHS Stat-compiler and Bangladesh MICS 2009
According to a recent publication “Achieving the MDG’s with Equity” the ratio of the % # ! `" countries and also in South Asia. !
! % rates (U5MR) among selected Asian countries, as shown in Table 1. Rural U5MR is 25% higher, a ratio of U5MR , rural areas to urban areas, is therefore of 1.25.
rural areas The same relationship between urban and rural area is found for Bangladesh when a composite deprivation index (CDI)13 is used (Table 2).
13 Composite deprivation index, based upon four indicators: net attendance ratio in secondary school, sanitation coverage, adult literacy and skilled attendance at birth.
11
Table 2: Comparisons of urban and rural areas by ranking in the composite deprivation index
Bangladesh
Total CDI
Urban CDI
Rural CDI
Rural % higher:
55
45
58
29%
Urban-rural differences for MICS 2009 indicators When analysing the components of the CDI, the most obvious observation is the absence of any difference between urban and rural areas in access to improved sanitation ; !
